"Steinar H. Gunderson" <[EMAIL PROTECTED]> writes: > Please try the attached patch; I don't have an EM64T, but it looks > like everything but the hand-created SSE2 assembly should be OK. > (IOW, the ATL_3DNow_* flags aren't set for amd64 in -base.) I don't > really have an idea of what would be the most appropriate prefetch > instruction, but prefetchnta seems to have been used other places, > so here goes:
Works for me; thanks for picking up where I, in turn, left off. ;-) Just one nit: the first hunk for ATL_dmm14x1x56_sse2pABC_K.c uses prefetcht0 rather than prefetchnta; while it runs fine regardless (prefetch instructions being essentially optional AIUI and therefore an ironic source of illegal instruction errors), you should probably still use the same alternative throughout. Anyway, thanks again for the patch. -- Aaron M. Ucko, KB1CJC (amu at alum.mit.edu, ucko at debian.org) Finger [EMAIL PROTECTED] (NOT a valid e-mail address) for more info. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
